混合研究视角下的演讲稿情感分析——以外研社•国才杯全国英语演讲大赛冠军为例
Sentiment Analysis of Speech Drafts from the Perspective of a Mixed Method —A Case Study of the Champion of “FLTRP Cup” English Public Speaking Contest
摘要: 本研究采用混合研究方法，以2021年外研社•国才杯全国大学生英语演讲大赛冠军决赛过程中的四篇演讲稿为例，进行情感值分句与分段统计、情感变化趋势分析、8类情绪词分类编码统计。研究结果显示，在四篇演讲稿中，正向情绪词明显多于负向情绪词；情感值的至高点出现在演讲的中后段；演讲结束时，情感值明显高于开头情感值。分段情感值波动较大，但总体呈上升趋势。量化与质性分析结果显示，8类情绪词各自占比与占比高低排序存在较大差异。
Abstract: From the perspective of a mixed method, this study takes 4 final speech drafts of the champion in the 2021 “FLTRP Cup” National English Speech Contest for college students as examples, and conducts clause and segment statistics of sentiment value, analysis of the dynamic process of the fluctuation of sentiment, and classification and coding statistics of 8 categories of sentiment words. The results showed that in the four speeches, positive words outweigh negative words; the peak of the sentiment value mostly appears in the middle or at the end of a speech; the sentiment value at the beginning of a speech is much higher than that of the end. The sentiment value fluctuates dramatically on the paragraph level but is generally on the increase. Quantitative and qualitative analysis results showed that the frequency and the ranking of frequency of the 8 categories of the emotion words vary significantly.
文章引用：汤慧桃. 混合研究视角下的演讲稿情感分析——以外研社•国才杯全国英语演讲大赛冠军为例[J]. 现代语言学, 2023, 11(5): 2121-2133. https://doi.org/10.12677/ML.2023.115288

参考文献

[1] Yang, L.Y., et al. (2021) Exploring the Efficacy of Automatically Generated Counterfactuals for Sentiment Analysis. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, Vol. 1, 1-6 August 2021, 306-316. [Google Scholar] [CrossRef] 
[2] 张博, 董瑞海. 自然语言处理技术赋能教育智能发展——人工智能科学家的视角[J]. 华东师范大学学报(教育科学版), 2022, 40(9): 19-31.
[3] 人工智能发展报告(2011-2020) [R]. 清华大学人工智能研究院, 2020.
[4] Zheng, Y.W., Zhang, R.C., Mensah, S. and Mao, Y.Y. (2020) Replicate, Walk, and Stop on Syntax: An Effective Neural Network Model for Aspect-Level Sentiment Classification. Proceedings of the Thirty-Fourth AAAI Conference on Artificial Intelligence, 34, 9685-9692. [Google Scholar] [CrossRef] 
[5] Newman, H. and Joyner, D. (2018) Sentiment Analysis of Student Evaluations of Teaching. International Conference on Artificial Intelligence in Education, Vol. 10948, Springer, Cham, 246-250. [Google Scholar] [CrossRef] 
[6] Rajput, Q., et al. (2016) Lexicon-Based Sentiment Analysis of Teachers’ Evaluation. Applied Computational Intelligence and Soft Computing, 2016, Article ID: 2385429. [Google Scholar] [CrossRef] 
[7] Mohammad, S. and Turney, P. (2010) Emotions Evoked by Common Words and Phrases: Using Mechanical Turk to Create an Emotion Lexicon. CAAGET’10: Proceedings of the NAACL HLT 2010 Workshop on Computational Approaches to Analysis and Generation of Emotion in Text, Los Angeles, CA, June 2010, 26-34.